Faculty Examples: Build Metacognitive Habits

Create intentional moments for students to reflect on their own thinking and learning processes, with or without AI.

Metacognition, thinking about one's own thinking, is what turns AI use into a learning opportunity rather than a shortcut around it. When students are asked to notice and name their own reasoning, they build a habit that serves them whether or not AI is part of a given task.

Below are examples from UO faculty.

Professor Brown encourages his students to confront what AI thinks their voice is, then argue about what's actually theirs. Prof. Brown says, "I would be thrilled if many students decided as a result of this class that they do not need to rely on AI in their writing processes."

Design Moves

• Course policy draws a hard line at the creative act itself: AI is barred from writing stories, generating scenes, imitating style, giving workshop feedback, or rewriting prose, but permitted for research, broad brainstorming, and analytical critique.
• Signature task: students ask AI to analyze their own writing style and name authors with a similar voice, then the class discusses what's actually unique about each student's writing, using AI's likely generic characterization as a foil for genuine self-reflection.
• Runs on a mix of in-class and take-home work based on trust and evidence, not lockdown conditions.

In Professor Heinz's HIST 308 course, the essay's final body paragraph requires evaluative judgment: which historical change mattered most, which mattered least, and why, including citing at least one example the student admits is genuinely unclear or complicated. Moreover, students have been reflecting on points of uncertainty across the essay's revision cycle. 

Design Moves

• Essay Paragraph 4 ("Evaluation") requires students to argue which historical change was most and least significant and to include one "complicated or unclear example," prompting reflection and judgment rather than simple summary.
• Process checkpoints (thesis draft, three annotated timeline points, essay outline) each require a field for "What I'm unsure about." Uncertainty is built into the submission itself, not just invited at the end.
• Final reflection (300+ words) requires quoting actual feedback received from their GE and describing how they responded to it in revision, naming a specific external input and their response to it, not just general self-assessment.
• A required question asks students to trace how their thesis evolved from their first draft, making revision itself part of what's being reflected on.

Professor Francese describes the broader purpose of his classes as foundational knowledge, understanding conceptual frameworks, and legal literacy, and builds a required reflection into the final paper on that basis. Prof. Francese says, "Outside of a well-designed AI tutor, I have yet to see an AI tool or an AI-based assignment that would help achieve these broader learning goals."

Design Moves

• Students choose their own event and outlets to compare, which increases motivation and makes the work harder to outsource.
• Clear, staged steps: conceptualize, outline, then conduct the comparative analysis.
• Requires annotation and screenshots documenting how sources were integrated, as evidence of process.
• A written reflection on the research and writing process accompanies the final paper, tied to a topic the student chose themselves, not a generic prompt.

Reflection runs continuously through Professor Jaramillo's course rather than as a single end-of-term assignment. Journaling, video, and peer review are all standing features, not one-off checkpoints.

Design Moves

• Requires AI-interaction journals kept throughout the term, documenting ongoing use rather than a single retrospective account.
• Students' weekly videos give reflection a recurring cadence, not a one-time deliverable.
• Peer review and process documentation are built in alongside these journals and video reflection components.
• Students also analyze AI's own output for bias, exclusions, and what it flattens or amplifies. In other words, reflection extends to evaluating the tool, not just their own process.